{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,24]],"date-time":"2026-10-24T00:29:16Z","timestamp":1792801756465,"version":"4.5.0"},"reference-count":26,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T00:00:00Z","timestamp":1761868800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100021856","name":"Ministero dell'Universit\u00e0 e della Ricerca","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100021856","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Hum. Neurosci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>In neuroscience, the muscle synergy method is a widely known computational approach for studying motor control from electromyographic (EMG) recordings. Standard algorithms for synergy extraction rely on a linearity assumption for synergy combination. However, the interactions between muscle groups and movement dynamics often exhibit non-linear characteristics, suggesting the need for alternative approaches. In this context, autoencoders (AEs) have been proposed as promising tools. However, previous studies focused on the reconstruction accuracy optimization and not on the structure of the synergies, and the influence of AE design parameters has not been thoroughly investigated. This study aims to explore the impact of different activation functions on the effectiveness of AEs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>To this end, we used a rich dataset of upper-limb EMG signals recorded from 16 muscles in 15 participants performing reaching movements toward 9 targets across 5 planes. We evaluated the effects of combining four activation functions in the encoder and decoder layers\u2014linear, ReLU, sigmoid, and tanh\u2014and compared to standard non-negative matrix factorization (NMF).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Our findings show that the extracted synergies are highly sensitive to the AE architecture. Notably, the configurations obtaining the best signal reconstruction do not correspond to the most physiologically meaningful synergies, which were instead achieved with the ReLU+tanh configuration.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>This suggests that optimizing reconstruction accuracy may result in non-interpretable synergy structures. This research emphasizes the role of non-linear techniques in extracting muscle synergy from different datasets (e.g., lower limbs, full-body movements, patient populations) and identifies the optimal combination of transfer functions for the encoder and decoder layers.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnhum.2025.1699799","type":"journal-article","created":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T06:23:44Z","timestamp":1761891824000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["The trade-off between maximizing reconstruction and physiological interpretation of muscle synergies with autoencoders"],"prefix":"10.3389","volume":"19","author":[{"given":"Cristina","family":"Brambilla","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicol","family":"Moscatelli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Valentina","family":"Lanzani","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lorenzo","family":"Molinari Tosatti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alessandro","family":"Brusaferri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alessandro","family":"Scano","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,10,31]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"494","DOI":"10.3389\/fphys.2014.00494","article-title":"Detection of the electromechanical delay and its components during voluntary isometric contraction of the quadriceps femoris muscle.","volume":"5","author":"Begovic","year":"2014","journal-title":"Front. Physiol."},{"key":"B2","doi-asserted-by":"publisher","first-page":"437","DOI":"10.1177\/107385802236969","article-title":"Modular organization of spinal motor systems.","volume":"8","author":"Bizzi","year":"2002","journal-title":"Neuroscientist"},{"key":"B3","doi-asserted-by":"publisher","first-page":"8584","DOI":"10.3390\/s22228584","article-title":"The number and structure of muscle synergies depend on the number of recorded muscles: A pilot simulation study with opensim.","volume":"22","author":"Brambilla","year":"2022","journal-title":"Sensors"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/IJCNN.2019.8851975","article-title":"An undercomplete autoencoder to extract muscle synergies for motor intention detection","author":"Buongiorno","year":"2019","journal-title":"Proceedins of the 2019 international joint conference on neural networks (IJCNN)"},{"key":"B5","doi-asserted-by":"publisher","first-page":"219","DOI":"10.3390\/info11040219","article-title":"Task-oriented muscle synergy extraction using an autoencoder-based neural model.","volume":"11","author":"Buongiorno","year":"2020","journal-title":"Information"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1580","DOI":"10.1152\/jn.00625.2019","article-title":"Approaches to revealing the neural basis of muscle synergies: A review and a critique.","volume":"125","author":"Cheung","year":"2021","journal-title":"J. Neurophysiol."},{"key":"B7","doi-asserted-by":"publisher","first-page":"844","DOI":"10.1152\/jn.00825.2009","article-title":"Merging of healthy motor modules predicts reduced locomotor performance and muscle coordination complexity post-stroke.","volume":"103","author":"Clark","year":"2010","journal-title":"J. Neurophysiol."},{"key":"B8","doi-asserted-by":"publisher","first-page":"7791","DOI":"10.1523\/JNEUROSCI.0830-06.2006","article-title":"Control of fast-reaching movements by muscle synergy combinations.","volume":"26","author":"d\u2019Avella","year":"2006","journal-title":"J. Neurosci."},{"key":"B9","doi-asserted-by":"publisher","first-page":"231","DOI":"10.1007\/978-981-15-5093-5_22","article-title":"A nonlinear autoencoder for kinematic synergy extraction from movement data acquired with HTC vive trackers","author":"De Feudis","year":"2021","journal-title":"Progresses in artificial intelligence and neural systems"},{"key":"B10","doi-asserted-by":"publisher","first-page":"2882","DOI":"10.3390\/s21082882","article-title":"Online adaptive prediction of human motion intention based on sEMG.","volume":"21","author":"Ding","year":"2021","journal-title":"Sensors"},{"key":"B11","doi-asserted-by":"publisher","first-page":"107940","DOI":"10.1016\/j.bspc.2025.107940","article-title":"On autoencoders for extracting muscle synergies: A study in highly variable upper limb movements.","volume":"108","author":"Giraud","year":"2025","journal-title":"Biomed. Signal Processing Control"},{"key":"B12","doi-asserted-by":"publisher","first-page":"5269","DOI":"10.1523\/JNEUROSCI.5626-03.2004","article-title":"Modular premotor drives and unit bursts as primitives for frog motor behaviors.","volume":"24","author":"Hart","year":"2004","journal-title":"J. Neurosci."},{"key":"B13","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1016\/S1050-6411(00)00027-4","article-title":"Development of recommendations for SEMG sensors and sensor placement procedures.","volume":"10","author":"Hermens","year":"2000","journal-title":"J. Electromyogr. Kinesiol."},{"key":"B14","doi-asserted-by":"publisher","first-page":"573","DOI":"10.1152\/jn.01054.2007","article-title":"A simple experimentally based model using proprioceptive regulation of motor primitives captures adjusted trajectory formation in spinal frogs.","volume":"103","author":"Kargo","year":"2010","journal-title":"J. Neurophysiol."},{"key":"B15","doi-asserted-by":"publisher","first-page":"225","DOI":"10.1016\/j.gaitpost.2023.10.009","article-title":"Characteristics of muscle synergy extracted using an autoencoder in patients with stroke during the curved walking in comparison with healthy controls.","volume":"107","author":"Lee","year":"2024","journal-title":"Gait Posture"},{"key":"B16","doi-asserted-by":"publisher","first-page":"352","DOI":"10.1016\/S0006-8993(99)02298-2","article-title":"Deficits in the coordination of agonist and antagonist muscles in stroke patients: Implications for normal motor control.","volume":"853","author":"Levin","year":"2000","journal-title":"Brain Res."},{"key":"B17","doi-asserted-by":"publisher","first-page":"R87651","DOI":"10.7554\/eLife.87651","article-title":"Dissecting muscle synergies in the task space.","volume":"12","author":"O\u2019Reilly","year":"2024","journal-title":"eLife"},{"key":"B18","doi-asserted-by":"publisher","first-page":"6284","DOI":"10.1109\/ICASSP40776.2020.9054734","article-title":"F0-consistent many-to-many non-parallel voice conversion via conditional autoencoder","author":"Qian","year":"2020","journal-title":"Proceedings of the ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"},{"key":"B19","doi-asserted-by":"publisher","first-page":"24","DOI":"10.3389\/fncom.2014.00024","article-title":"Dimensionality of joint torques and muscle patterns for reaching.","volume":"8","author":"Russo","year":"2014","journal-title":"Front. Comput. Neurosci."},{"key":"B20","doi-asserted-by":"publisher","first-page":"726","DOI":"10.1016\/j.neunet.2021.09.014","article-title":"ARAE: Adversarially robust training of autoencoders improves novelty detection.","volume":"144","author":"Salehi","year":"2021","journal-title":"Neural Networks"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1231","DOI":"10.3389\/fphys.2019.01231","article-title":"A comprehensive spatial mapping of muscle synergies in highly variable upper-limb movements of healthy subjects.","volume":"10","author":"Scano","year":"2019","journal-title":"Front. Physiol."},{"key":"B22","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1152\/jn.00379.2021","article-title":"Mixed matrix factorization: A novel algorithm for the extraction of kinematic-muscular synergies.","volume":"127","author":"Scano","year":"2022","journal-title":"J. Neurophysiol."},{"key":"B23","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1007\/978-3-319-44781-0_6","article-title":"Extracting muscle synergy patterns from EMG Data using autoencoders","author":"Sp\u00fcler","year":"2016","journal-title":"Artificial Neural Networks and Machine Learning \u2013 ICANN 2016"},{"key":"B24","doi-asserted-by":"publisher","first-page":"105","DOI":"10.3389\/fncom.2013.00105","article-title":"The number and choice of muscles impact the results of muscle synergy analyses.","volume":"7","author":"Steele","year":"2013","journal-title":"Front. Comput. Neurosci."},{"key":"B25","doi-asserted-by":"publisher","first-page":"2199","DOI":"10.1152\/jn.00222.2005","article-title":"Matrix factorization algorithms for the identification of muscle synergies: Evaluation on simulated and experimental data sets.","volume":"95","author":"Tresch","year":"2006","journal-title":"J. Neurophysiol."},{"key":"B26","doi-asserted-by":"publisher","first-page":"732156","DOI":"10.3389\/fnins.2022.732156","article-title":"Evaluation of methods for the extraction of spatial muscle synergies.","volume":"16","author":"Zhao","year":"2022","journal-title":"Front. Neurosci."}],"container-title":["Frontiers in Human Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnhum.2025.1699799\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T06:23:44Z","timestamp":1761891824000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnhum.2025.1699799\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,31]]},"references-count":26,"alternative-id":["10.3389\/fnhum.2025.1699799"],"URL":"https:\/\/doi.org\/10.3389\/fnhum.2025.1699799","relation":{},"ISSN":["1662-5161"],"issn-type":[{"value":"1662-5161","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,31]]},"article-number":"1699799"}}